透過您的圖書館登入
IP:184.72.135.210
  • 學位論文

探尋精神分裂症的易感性基因:位置選殖策略

Searching for susceptibility genes of schizophrenia: positional cloning strategy

指導教授 : 胡海國
共同指導教授 : 尹相姝(Hsiang-Shu Yin)

摘要


精神分裂症是一嚴重慢性的神經精神疾病,已知遺傳病因扮演致病重要角色,目前致病模式傾向由多個基因與多種環境因素交互作用而成,是一種複雜疾病 (complex disorder)。探尋這類疾病易感性基因的方法主要為位置選殖策略 (positional cloning strategy),包括遺傳連鎖研究 (genetic linkage study),候選基因關聯性研究 (candidate gene association study),位置候選基因研究 (positional candidate gene association study),全基因體關聯性研究 (genome-wide association study)。本論文總結作者過去以位置選殖的方法,以台灣精神分裂症家族為樣本所做的易感性基因探尋的研究結果,並討論各種策略的優缺點,及未來的展望。 關於遺傳連鎖研究,以第八對染色體短臂22-12區域 (8p22-12) 的遺傳連鎖研究為例。本研究針對52個至少有兩名罹病手足的臺灣精神分裂症患者家族,使用染色體 8p22-21上的11個微衛星標記 (microsatellite markers),研究這個染色體區域是否與精神分裂症有連鎖證據。結果發現於標記 D8S1222 發現最大無母數連鎖分數 (NPL分數) 為2.45 (p = 0.008)。標記D8S1222距離GGF2基因 (glial growth factor 2) 之外顯子 (exon) 1約 400 kbp,而GGF2是NRG1基因 (Neuregulin 1) 的同質體 (isoform),是可能性極高的精神分裂症候選基因之一。本研究結果提供了臺灣人樣本之精神分裂症與位於染色體 8p21,接近NRG1基因位點建議性的連鎖證據。 關於候選基因關聯性研究,以dystrobrevin-binding protein 1 (DTNBP1) 的候選基因關聯性研究為例。許多連鎖研究發現精神分裂症與第六對染色體短臂 (6p) 區域有顯著的連鎖證據,其中包括候選基因DTNBP1。本研究針對693個至少有兩名罹病手足的臺灣精神分裂症患者家族,探討此疾病與DTNBP1的關聯性證據。定出此基因之9個單核苷酸多型性變異 (single nucleotide polymorphism, SNP) 的基因型,標記間平均距離為17 kb。使用GOLD計算標記間的連鎖不平衡 (linkage disequilibrium, LD),並使用TRANSMIT進行單一基因位點與單套型 (haplotype)關聯性分析。結果發現單一基因位點與單套型分析中,精神分裂症與DTNBP1皆無顯著關聯性。本研究以相當大的單一種族樣本,仍無法重現之前其他研究中DTNBP1與精神分裂症的關聯性,此基因在臺灣家族樣本之精神分裂症病因中可能並未扮演關鍵角色。 關於位置候選基因關聯性研究,我們以全台灣收集的557個精神分裂症家族樣本進行全基因體遺傳連鎖研究 (genome-wide linkage study) 發現10q22連鎖的證據,進一步使用原先的家族樣本,標認出這個染色體區域的易感性基因。研究樣本為476個精神分裂症家庭,每個家庭至少有兩位或以上的手足罹患精神分裂症,這些家庭中至少有一位患病手足接受持續性表現測驗 (Continuous Performance Test, CPT) 以及威斯康辛卡片分類測驗 (Wisconsin Card Sorting Test, WCST)。 第一階段,首先我們以GENEFINDER在之前全基因體遺傳分析樣本中標認出遺傳連鎖證據的95%信賴區間,位於標記D10S1432 (93.97cM) 和D10S1239 (121.81cM) 之間,在這個區間選取額外的18個微衛星標記,定出476個家庭樣本的基因型,使用MERLIN作連鎖分析,發現最大的NPL分數為2.79落在標記D10S195 (94.72cM) 上。其次,以階層群集分析 (hierarchical clustering),依據患病手足的CPT和WCST的表現,將這些家庭分成四個亞群,連鎖分析發現在持續注意力及執行功能皆缺損的亞群家庭樣本中 (Attention Deficit and Execution Deficit, ADED),其最大的NPL分數達到3.70 (p = 0.00008),也落在標記D10S195上。 第二階段,我們在標記D10S1432 (93.97 cM) and D10S580 (95.52cM)之間選取了79個單套型標記 (haplotype tagged) SNP,定出90個ADED家庭的基因型,使用FBAT和TRANSMIT分析發現有9個SNP有顯著關聯性,使用最長連續顯著段落分析 (the longest significance run, LSR) 發現橫跨四個SNP (rs4492736, rs7087762, rs12644, rs4746136) 的基因體區域與精神分裂症的關聯性達到顯著 (p = 0.0048)。這個區域長度為427kb,包含了9個基因。 第三階段,以即時反轉錄聚合脢鏈鎖反應 (real-time RT-PCR) 的方法,定出顯著關聯區域中的9個基因的傳訊核醣核酸 (mRNA) 在EBV轉型淋巴球 (EBV-transformed lymphocyte) 中的相對表現量,研究樣本為95位精神分裂症病患及36位正常對照個案。結果發現ANXA7 (annexin A7),PPP3CB (protein phosphatase 3 (formerly 2B), catalytic subunit, beta isoform),和DNAJC9 (DnaJ (Hsp40) homolog, subfamily C, member 9) 三個基因的mRNA的表現量在病患組顯著較低,而ZMYND17 (zinc finger, MYND-type containing 17) 的表現量在病患組顯著為高。帶有危險基因型的病患其ANXA7的表現量較沒有帶危險基因型的病患及正常對照組顯著為低。病患組的ANXA7和PPP3CB的表現量與WCST的表現呈現顯著的正相關。 這個研究發現ANXA7,PPP3CB, DNAJC9,ZMYND17與持續注意力及執行功能皆缺損的精神分裂症亞群有顯著的遺傳關聯性,且表現量在病患組與正常對照組也有顯著差異,以基因功能及基因表現與基因型及神經認知缺損的關係推論,ANXA7和PPP3CB是最可能的候選基因。其基因病理機轉可能與基因表現量的改變有關,進一步探尋在這兩個基因的調節區域 (regulatory element) 的功能性變異,以及基因在神經細胞及神經系統扮演的主要功能,是未來研究的重點。 綜合以上研究,結合遺傳連鎖研究找出的染色體位置,進行位置候選基因研究,可增加關聯性研究的事前機率 (prior probability),使用精神分裂症的神經生物缺損作為內生性表現型 (endophenotype),可減少精神分裂症表現型的異質性,再輔以基因表現的研究,找出最可能的候選基因,是較好的策略。未來的研究,包括全基因體關聯性研究,複製數量變異 (copy number variation),全外顯體定序研究 (exome sequencing) 及使用誘導多能幹細胞 (induced pluripotent stem cell) 進行基因及神經生物功能的研究。

並列摘要


Schizophrenia is a chronic debilitating neuropsychiatric illness. The causes of schizophrenia are still unclear, but family, twin, and adoption studies indicate that it has a strong genetic component and high heritability. It is a complex disorder. The methods for searching the susceptibility genes are positional cloning strategies, including genetic linkage study, candidate gene association study, positional candidate gene association study, and genome-wide association study. This thesis summarized the research results using positional cloning to search for the susceptibility genes of schizophrenia in Taiwanese family samples, and commented on the strength and weakness of different strategies and proposed future perspectives. As for the genetic linkage study, we took the linkage study on chromosome 8p22-21 in Taiwanese schizophrenia families for example.This study genotyped eleven microsatellite markers on chromosome 8p22-21 in 52 Taiwanese schizophrenic families with at least two affected siblings. Maximum non-parametric linkage scores (NPL score) of 2.45 (p = 0.008) were obtained for the marker D8S1222. The marker D8S1222 was about 400 kbp distal to the exon 1 of glial growth factor 2 (GGF2), an isoform of Neuregulin 1 gene (NRG1), which has been highly suggested to be a candidate gene for schizophrenia. The results provide suggestive linkage evidence of schizophrenia to loci near NRG1 on chromosome 8p21 in an ethnically distinct Taiwanese sample. Further exploration of the candidate gene and nearby chromosome regions is warranted. As for the candidate gene association study, we took the association study between Dystrobrevin-binding protein 1 (DTNBP1) and schizophrenia in Taiwan. The aim was to examine the association evidence of the candidate gene in 693 Taiwanese families with at least two affected siblings of schizophrenia. We genotyped nine SNPs of this gene with average intermarker distance of 17 kb. Single locus and haplotype association analyses were performed with TRANSMIT program. We found no significant association between schizophrenia and DTNBP1 either through single locus or haplotype analyses. We failed to replicate the association evidence between DTNBP1 and schizophrenia and this gene may not play a major role in the etiology of schizophrenia in this Taiwanese family sample. As for the positional candidate gene association study, the genome wide linkage study of 557 family samples in Taiwan has revealed linkage evidence of chromosome 10q22 to schizophrenia. We further performed fine mapping study to identify the susceptibility genes in this region. The study sample comes from the original family sample, including 476 Han Taiwanese families having at least two siblings affected with schizophrenia and at least one of the affected siblings has received the Continuous Performance Test (CPT) and Wisconsin Card Sorting Test (WCST). First, based on the results of the previous linkage analysis, we identified the region from D10S1432 (93.97 cM) to D10S1239 (121.81 cM) as the region with a 95% confidence interval of linkage. Then we selected 18 additional microsatellite markers in this region for further genotyping. The maximum NPL score was 2.79 on D10S195 (94.72cM) using MERLIN. We clustered families into four different subgroups by performances of CPT and WCST of affected siblings using hierarchical clustering method. We found the maximum NPL score was 3.70 (p = 0.00008) on D10S195 in the family cluster with attention deficit and execution deficit (ADED). Second, we genotyped 79 haplotype tagSNPs between D10S1432 (93.97 cM) and D10S580 (95.52 cM) in 90 ADED families. Single-point association analysis using FBAT and TRANSMIT indicated significant transmission distortion for nine SNPs. Using the longest significance run (LSR) method, we identified a 427-kb genomic region that spanned the four SNPs (rs4492736, rs7087762, rs12644, rs4746136) as a significant candidate region (p = 0.0048). This genomic region encompasses nine genes. Third, we studied the relative mRNA transcript levels of the nine genes in the EBV-transformed lymphocytes of 95 schizophrenic patients and 36 normal controls. In schizophrenic patients, there was significantly lower expression of annexin A7 (ANXA7), protein phosphatase 3 (formerly 2B), catalytic subunit, beta isoform (PPP3CB), and DnaJ (Hsp40) homolog, subfamily C, member 9 (DNAJC9) and significantly higher expression of zinc finger, MYND-type containing 17 (ZMYND17). Expression of ANXA7 was significantly lower in patients with the risk allele of the tagSNP (rs12258241) than those without this allele and the controls. The expression levels of ANXA7 and PPP3CB are significantly positively correlated with the performance of WCST of the schizophrenic patients. We have identified ANXA7, PPP3CB, DNAJC9, and ZMYND17 as potential candidate genes for schizophrenia, especially in patients with deficits in sustained attention and executive function. Considering the biological function and the relationship with the neurocognitive deficits of these four genes, ANXA7 and PPP3CB are the most likely susceptibility genes. In summary, it is a better strategy to combine positional candidate gene association study and endophenotype approach using neurobiological deficits, and further to integrate the genetic association and gene expression results to identify the most potential candidate genes. In the future, genome-wide association study, copy number variations study, exome sequencing study and genetic and neurobiological studies using induced pluripotent stem cells may cast much insight for understanding the pathophysiological mechanism of this disorder.

參考文獻


Tien AY, Spevack TV, Jones DW, Pearlson GD, Schlaepfer TE, Strauss ME. Computerized Wisconsin Card Sorting Test: comparison with manual administration. Kaohsiung J Med Sci. 1996;12(8):479-85.
Hwu HG. Psychiatric diagnostic assessment. 2 ed. Taipei: Publication Committee, College of Medicine, National Taiwan University 1999.
Abecasis GR, Cardon LR, Cookson WO. A general test of association for quantitative traits in nuclear families. Am J Hum Genet. 2000;66(1):279-92.
Abecasis GR, Cookson WO. GOLD--graphical overview of linkage disequilibrium. Bioinformatics. 2000;16(2):182-3.
Abecasis GR, Cherny SS, Cookson WO, Cardon LR. Merlin--rapid analysis of dense genetic maps using sparse gene flow trees. Nat Genet. 2002;30(1):97-101.

延伸閱讀